Over the past decade, approaches to the diagnosis of dementia have subtly shifted from one in which various disorders are "ruled out" to one in which characteristic findings may be used to support a specific dementia diagnosis. This transition has in large part been fueled by advances in brain imaging that show relatively specific molecular and biochemical changes in Alzheimer's disease (AD) using positron emission tomography (PET), and characteristic anatomical changes using magnetic resonance imaging (MRI). The recent development of a PET imaging compound that labels amyloid, N-methyl [11C]2-(4'- methylaminophenyl)-6-hydroxybenzothiazole, nicknamed "PIB" for Pittsburgh compound B, is another major advance for the field that may help differentiate different causes of dementia. In particular, a relatively common group of degenerative dementias collectively known as frontotemporal lobar degeneration (FTLD) may be difficult to differentiate from AD during life. We hypothesize that the combination of PIB imaging along with traditional glucose metabolic PET imaging with [18F]Fluorodeoxyglucose (FDG) will effectively differentiate the two disorders. We plan to recruit a group of 45 patients with AD and 45 patients with FTLD through the Memory and Aging center at the University of California San Francisco, a major center for the study of AD and FTLD. Individuals will receive an intensive clinical diagnostic evaluation including cognitive testing, genetic testing, and MR imaging. They will be followed in a longitudinal cohort with a brain donation program. Subjects will be studied with PIB-PET and FDG-PET and both qualitative ratings and quantitative measurements of PIB and FDG uptake will be utilized. These measures will be compared to clinical diagnosis and, over the long term, autopsy diagnoses to define the sensitivity, specificity, and discriminative ability of the two techniques relative to the highest quality clinical diagnosis possible, and to one another. The significance of this project is that there are no current laboratory tests to differentiate these two dementing disorders. PIB could become a diagnostic tool that is useful in diagnosing AD and FTLD during life. This will be especially important as new treatments for both disorders are developed.